Molecular mechanism of heterogeneous ice nucleation on potassium feldspar

Pablo M. Piaggi; Wanqi Zhou

arxiv: 2511.13402 · v3 · submitted 2025-11-17 · ⚛️ physics.chem-ph

Molecular mechanism of heterogeneous ice nucleation on potassium feldspar

Wanqi Zhou , Pablo M. Piaggi This is my paper

Pith reviewed 2026-05-17 21:14 UTC · model grok-4.3

classification ⚛️ physics.chem-ph

keywords ice nucleationpotassium feldsparK-feldsparheterogeneous nucleationmolecular dynamics simulationmineral dustatmospheric ice formationcubic ice

0 comments

The pith

The (110) surface of potassium feldspar structures water to match cubic ice and nucleates ice most effectively at defects.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper uses machine-learning molecular dynamics to examine ice nucleation on different surfaces of potassium feldspar. It establishes that the (110) surface, when exposed at steps or other defects, provides the best template for ice formation because it arranges water molecules in a pattern similar to the (110) face of cubic ice. A sympathetic reader would care because this explains the high ice-nucleating ability of feldspar minerals in the atmosphere, which plays a key role in cloud formation and Earth's climate. The work moves from general recognition of feldspar's importance to a specific molecular mechanism involving particular crystal planes.

Core claim

Using machine-learning molecular dynamics simulations, the study identifies the (110) surface of K-feldspar, exposed at defects such as steps, as the most active plane for ice formation. This surface uniquely structures interfacial water into an arrangement resembling that on the (110) surface of cubic ice, providing an optimal template for nucleation. Advanced sampling directly observes the formation of clusters with cubic-ice structure whose orientation agrees with experiment.

What carries the argument

The (110) surface of potassium feldspar as a structural template that organizes interfacial water into a cubic ice (110)-like arrangement.

If this is right

K-feldspar particles with more exposed (110) surfaces or defects will nucleate ice at higher temperatures or lower supersaturations.
The dominance of feldspar in atmospheric ice nucleation is due to the prevalence of (110) exposures at defects rather than the (100) surface.
Molecular simulations can now predict ice nucleation efficiency on other mineral surfaces based on similar structural matching.
Atmospheric models can incorporate this specific surface-dependent mechanism for more accurate cloud physics.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the templating holds, controlling surface defects on particles could modulate their ice-nucleating activity in experiments.
This finding connects to broader questions of how crystal lattice matching influences heterogeneous nucleation in other systems like salt crystallization.
Testable extension: Prepare single-crystal feldspar samples with specific facets and measure nucleation rates to confirm preference for (110).

Load-bearing premise

The machine-learned interatomic potential used in the simulations accurately reproduces the real interactions between water and potassium feldspar surfaces at atmospheric temperatures.

What would settle it

An experiment showing that ice does not preferentially form on (110) surfaces or steps of feldspar particles, or that the nucleated ice does not have the predicted orientation matching cubic ice (110).

read the original abstract

Mineral dust aerosols strongly influence Earth's climate by acting as ice-nucleating particles (INPs). Feldspar minerals, particularly K-feldspar, are recognized as dominant INPs, and a previous study attributed this behavior to (100) surfaces exposed at defects. Using machine-learning molecular dynamics simulations, we systematically investigate ice nucleation on multiple K-feldspar surfaces. We identify the (110) surface, exposed at defects such as steps, as the most active plane for ice formation. This surface uniquely structures interfacial water into an arrangement resembling that on the (110) surface of cubic ice, providing an optimal template for nucleation. Using advanced sampling, we directly observe the formation of clusters with cubic-ice structure and their orientation agrees with experiment. These results provide a molecular-level explanation of how ice forms in our planet's atmosphere.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This paper uses ML molecular dynamics to shift the active ice-nucleating surface on K-feldspar from (100) to (110) at defects, with direct observation of cubic-ice-like water clusters.

read the letter

The central claim is that the (110) surface, exposed at steps and defects, templates water into a cubic-ice-like arrangement and is therefore the dominant plane for nucleation on K-feldspar. This differs from earlier work that highlighted the (100) surface. The simulations systematically compare several surfaces and use advanced sampling to watch cluster formation, with the resulting orientation said to match experiment. That supplies a molecular picture for an atmospherically important INP and is the clearest new element here. The comparison across surfaces and the focus on defect structures are done cleanly enough to be useful. The work engages prior literature on feldspar nucleation without obvious circularity. The main limitation is the machine-learned potential. The ordering of nucleation activity rests on accurate relative free energies for water structuring and hydrogen-bond networks at the mineral interface. The abstract gives no explicit DFT benchmarks or experimental checks for those interface quantities, so any systematic bias in ion-water or surface-water forces could invert the (110) versus (100) ranking. Convergence details and error estimates on the observed clusters are also not visible in the summary. This is a moderate rather than fatal gap, but it directly affects the strength of the central conclusion. The paper is aimed at atmospheric chemists and climate modelers who need better molecular constraints on ice nucleation. Readers who work with mineral surfaces or ML potentials will get the most out of the surface comparisons. It is coherent on its own terms and shows honest engagement with the problem, so it deserves a serious referee even though revisions on validation will likely be required. I would send it to peer review with a request for potential benchmarks and sampling statistics.

Referee Report

2 major / 2 minor

Summary. The paper uses machine-learning molecular dynamics simulations to study heterogeneous ice nucleation on potassium feldspar surfaces. It concludes that the (110) surface, exposed at defects such as steps, is the most active for ice formation due to its unique structuring of interfacial water into a cubic-ice-like arrangement on its (110) plane. Advanced sampling techniques allow direct observation of cubic-ice clusters, with orientations matching experimental findings, offering a molecular explanation for K-feldspar's role as an ice-nucleating particle in the atmosphere.

Significance. If the results hold, this work is significant for atmospheric science and physical chemistry as it provides a detailed molecular mechanism for ice nucleation on a key mineral dust component, potentially explaining its high efficiency as an INP. The use of advanced sampling to directly observe nucleation events and the agreement with experiment on cluster orientation are notable strengths. It challenges earlier attributions to the (100) surface and highlights the role of specific defect-exposed planes. However, the findings' impact depends critically on the fidelity of the machine-learned potential.

major comments (2)

Methods section on potential development: The manuscript lacks explicit benchmarks of the machine-learned interatomic potential against DFT calculations for key properties such as water adsorption energies, hydrogen bond strengths, or the relative stabilities of different surface terminations at the water-K-feldspar interface. As the central claim that the (110) surface provides an optimal template rests on the potential accurately ordering the nucleation free energy barriers across surfaces (see abstract and results on surface comparison), this omission is a load-bearing issue that could affect the reliability of the identified most active plane.
Results on advanced sampling: In the section describing the advanced sampling simulations, there are no reported details on convergence criteria, statistical error bars, or the number of independent runs for the observed formation of cubic-ice clusters. This makes it difficult to evaluate the robustness of the direct observation and the claimed agreement with experimental orientations.

minor comments (2)

Introduction: The reference to the previous study attributing activity to (100) surfaces should include the specific citation for clarity.
Figure captions: Some figure captions could be expanded to better describe the water structuring observed on different surfaces.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed review of our manuscript. We appreciate the emphasis on the importance of potential validation and sampling statistics for supporting our central claims. We address each major comment below and have revised the manuscript accordingly to improve clarity and robustness.

read point-by-point responses

Referee: Methods section on potential development: The manuscript lacks explicit benchmarks of the machine-learned interatomic potential against DFT calculations for key properties such as water adsorption energies, hydrogen bond strengths, or the relative stabilities of different surface terminations at the water-K-feldspar interface. As the central claim that the (110) surface provides an optimal template rests on the potential accurately ordering the nucleation free energy barriers across surfaces (see abstract and results on surface comparison), this omission is a load-bearing issue that could affect the reliability of the identified most active plane.

Authors: We agree that explicit interface-specific benchmarks strengthen the manuscript and directly support the reliability of the surface-dependent nucleation barrier ordering. Although the potential development section references prior DFT validation for bulk and clean surface properties, we have added a new subsection in the revised Methods with direct comparisons of water adsorption energies, interfacial hydrogen bond strengths, and relative stabilities of different K-feldspar surface terminations against DFT reference calculations. These benchmarks confirm that the ML potential reproduces the key energetic ordering that underpins the identification of the (110) surface as the most active template. revision: yes
Referee: Results on advanced sampling: In the section describing the advanced sampling simulations, there are no reported details on convergence criteria, statistical error bars, or the number of independent runs for the observed formation of cubic-ice clusters. This makes it difficult to evaluate the robustness of the direct observation and the claimed agreement with experimental orientations.

Authors: We thank the referee for noting this gap in reporting. In the revised manuscript we have expanded the advanced sampling results section to explicitly state the convergence criteria (based on stabilization of the ice-like order parameter and free-energy profiles), the method used to estimate statistical uncertainties (block averaging across trajectory segments), and the number of independent runs performed. These details confirm the reproducibility of the cubic-ice cluster formation and the consistency of the observed orientations with experiment. revision: yes

Circularity Check

0 steps flagged

No circularity: claim rests on direct simulation observations

full rationale

The paper derives its central claim—that the (110) surface is the most active for ice nucleation due to unique structuring of interfacial water into a cubic-ice-like arrangement—directly from machine-learning molecular dynamics simulations and advanced sampling that observe cluster formation and orientation. No step reduces by construction to a fitted parameter, self-definition, or load-bearing self-citation; the results are presented as computational observations that can be compared to external experiment. The derivation chain is therefore self-contained and independent of the target conclusion.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim depends on the transferability and accuracy of the machine-learned potential for the water-feldspar interface and on the assumption that the chosen simulation cells and sampling protocols capture the relevant nucleation pathways without significant finite-size or timescale artifacts.

axioms (2)

domain assumption Machine-learned potentials trained on DFT data faithfully reproduce the relevant water-mineral energetics and dynamics at the simulation conditions.
Invoked implicitly when using MLMD to observe nucleation events.
domain assumption The (110) surface at steps is the dominant defect-exposed plane under atmospheric conditions.
Used to link simulation results to real particle behavior.

pith-pipeline@v0.9.0 · 5431 in / 1294 out tokens · 32088 ms · 2026-05-17T21:14:05.244029+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We identify the (110) surface... structures interfacial water into an arrangement resembling that on the (110) surface of cubic ice... advanced sampling... Q6 Steinhardt order parameter... classical nucleation theory (CNT) fit
IndisputableMonolith/Foundation/AlexanderDuality.lean alexander_duality_circle_linking unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

MSE between water density profiles... minimum MSE corresponds to the feldspar (110)-α surface and the ice Ic (110) surface

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

29 extracted references · 29 canonical work pages

[1]

M¨ ulmenst¨adt, O

J. M¨ ulmenst¨adt, O. Sourdeval, J. Delano ¨e, J. Quaas, Frequency of occurrence of rain from liquid-, mixed-, and ice-phase clouds derived from A-Train satellite retrievals.Geophys. Res. Lett.42(15), 6502–6509 (2015)

work page 2015
[2]

Vergara-Temprado,et al., Strong control of Southern Ocean cloud reflectivity by ice- nucleating particles.Proc

J. Vergara-Temprado,et al., Strong control of Southern Ocean cloud reflectivity by ice- nucleating particles.Proc. Natl. Acad. Sci. U. S. A.115(11), 2687–2692 (2018)

work page 2018
[3]

Flato,et al., Evaluation of climate models, inClimate change 2013: the physical science basis

G. Flato,et al., Evaluation of climate models, inClimate change 2013: the physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change(Cambridge University Press), pp. 741–866 (2014)

work page 2013
[4]

D. A. Knopf, P. A. Alpert, Atmospheric ice nucleation.Nat. Rev. Phys.5(4), 203–217 (2023)

work page 2023
[5]

Hoose, O

C. Hoose, O. M ¨ohler, Heterogeneous ice nucleation on atmospheric aerosols: a review of results from laboratory experiments.Atmos. Chem. Phys.12(20), 9817–9854 (2012)

work page 2012
[6]

V. I. Kalikmanov, Classical nucleation theory, inNucleation theory(Springer), pp. 17–41 (2012)

work page 2012
[7]

J. D. Atkinson,et al., The importance of feldspar for ice nucleation by mineral dust in mixed- phase clouds.Nature498(7454), 355–358 (2013)

work page 2013
[8]

Yakobi-Hancock, L

J. Yakobi-Hancock, L. Ladino, J. Abbatt, Feldspar minerals as efficient deposition ice nuclei. Atmos. Chem. Phys.13(22), 11175–11185 (2013)

work page 2013
[9]

A. D. Harrison,et al., Not all feldspars are equal: a survey of ice nucleating properties across the feldspar group of minerals.Atmos. Chem. Phys.16(17), 10927–10940 (2016)

work page 2016
[10]

A. D. Harrison,et al., The ice-nucleating ability of quartz immersed in water and its atmospheric importance compared to K-feldspar.Atmos. Chem. Phys.19(17), 11343–11361 (2019)

work page 2019
[11]

Kiselev,et al., Active sites in heterogeneous ice nucleation—the example of K-rich feldspars

A. Kiselev,et al., Active sites in heterogeneous ice nucleation—the example of K-rich feldspars. Science355(6323), 367–371 (2017). 13

work page 2017
[12]

T. F. Whale,et al., The role of phase separation and related topography in the exceptional ice-nucleating ability of alkali feldspars.Phys. Chem. Chem. Phys.19(46), 31186–31193 (2017)

work page 2017
[13]

M. A. Holden,et al., High-speed imaging of ice nucleation in water proves the existence of active sites.Sci. Adv.5(2), eaav4316 (2019)

work page 2019
[14]

A. Soni, G. Patey, Simulations of water structure and the possibility of ice nucleation on selected crystal planes of K-feldspar.J. Chem. Phys.150(21) (2019)

work page 2019
[15]

E. Pach, A. Verdaguer, Pores dominate ice nucleation on feldspars.J. Phys. Chem. C123(34), 20998–21004 (2019)

work page 2019
[16]

A. A. Kiselev,et al., Effect of chemically induced fracturing on the ice nucleation activity of alkali feldspar.Atmos. Chem. Phys.21(15), 11801–11814 (2021)

work page 2021
[17]

Kumar, A

A. Kumar, A. K. Bertram, G. N. Patey, Molecular simulations of feldspar surfaces interact- ing with aqueous inorganic solutions: interfacial water/ion structure and implications for ice nucleation.ACS Earth Space Chem.5(8), 2169–2183 (2021)

work page 2021
[18]

Keinert, K

A. Keinert, K. Deck, T. Gaedeke, T. Leisner, A. A. Kiselev, Mechanism of ice nucleation in liquid water on alkali feldspars.Faraday Discuss.235, 148–161 (2022)

work page 2022
[19]

Franceschi,et al., How water binds to microcline feldspar (001).J

G. Franceschi,et al., How water binds to microcline feldspar (001).J. Phys. Chem. Lett.15(1), 15–22 (2023)

work page 2023
[20]

Dickbreder,et al., Atomic structure and water arrangement on K-feldspar microcline (001)

T. Dickbreder,et al., Atomic structure and water arrangement on K-feldspar microcline (001). Nanoscale16(7), 3462–3473 (2024)

work page 2024
[21]

P. M. Piaggi, J. Weis, A. Z. Panagiotopoulos, P. G. Debenedetti, R. Car, Homogeneous ice nucleation in an ab initio machine-learning model of water.Proc. Natl. Acad. Sci. U. S. A. 119(33), e2207294119 (2022)

work page 2022
[22]

A. Soni, G. Patey, Using machine learning with atomistic surface and local water features to predict heterogeneous ice nucleation.J. Chem. Phys.160(12) (2024). 14

work page 2024
[23]

A. Soni, G. Patey, Unraveling the mechanism of ice nucleation by mica (001) surfaces.J. Phys. Chem. C125(48), 26927–26941 (2021)

work page 2021
[24]

A. Soni, G. Patey, How microscopic features of mineral surfaces critically influence heteroge- neous ice nucleation.J. Phys. Chem. C125(19), 10723–10737 (2021)

work page 2021
[25]

Vonnegut, The nucleation of ice formation by silver iodide.J

B. Vonnegut, The nucleation of ice formation by silver iodide.J. Appl. Phys.18(7), 593–595 (1947)

work page 1947
[26]

M. Chen, L. Tan, H. Wang, L. Zhang, H. Niu, Exploring homogeneous ice nucleation through molecular dynamics simulations with ab initio accuracy.Phys. Rev. B:Condens. Matter Mater. Phys.111(13), 134116 (2025)

work page 2025
[27]

P. J. Steinhardt, D. R. Nelson, M. Ronchetti, Bond-orientational order in liquids and glasses. Phys. Rev. B:Condens. Matter Mater. Phys.28(2), 784 (1983)

work page 1983
[28]

P. M. Piaggi, A. Selloni, A. Z. Panagiotopoulos, R. Car, P. G. Debenedetti, A first-principles machine-learning force field for heterogeneous ice nucleation on microcline feldspar.Faraday Discuss.249, 98–113 (2024)

work page 2024
[29]

W. Zhou, P. M. Piaggi, Data from ”The (110) surface is the key to the high ice nucleation efficiency of potassium feldspar”, Zenodo (2025), doi:10.5281/zenodo.17610332,https: //doi.org/10.5281/zenodo.17610332. Acknowledgments We are grateful for the computational resources provided on the Hyperion cluster at the Donostia International Physics Center (DIPC...

work page doi:10.5281/zenodo.17610332 2025

[1] [1]

M¨ ulmenst¨adt, O

J. M¨ ulmenst¨adt, O. Sourdeval, J. Delano ¨e, J. Quaas, Frequency of occurrence of rain from liquid-, mixed-, and ice-phase clouds derived from A-Train satellite retrievals.Geophys. Res. Lett.42(15), 6502–6509 (2015)

work page 2015

[2] [2]

Vergara-Temprado,et al., Strong control of Southern Ocean cloud reflectivity by ice- nucleating particles.Proc

J. Vergara-Temprado,et al., Strong control of Southern Ocean cloud reflectivity by ice- nucleating particles.Proc. Natl. Acad. Sci. U. S. A.115(11), 2687–2692 (2018)

work page 2018

[3] [3]

Flato,et al., Evaluation of climate models, inClimate change 2013: the physical science basis

G. Flato,et al., Evaluation of climate models, inClimate change 2013: the physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change(Cambridge University Press), pp. 741–866 (2014)

work page 2013

[4] [4]

D. A. Knopf, P. A. Alpert, Atmospheric ice nucleation.Nat. Rev. Phys.5(4), 203–217 (2023)

work page 2023

[5] [5]

Hoose, O

C. Hoose, O. M ¨ohler, Heterogeneous ice nucleation on atmospheric aerosols: a review of results from laboratory experiments.Atmos. Chem. Phys.12(20), 9817–9854 (2012)

work page 2012

[6] [6]

V. I. Kalikmanov, Classical nucleation theory, inNucleation theory(Springer), pp. 17–41 (2012)

work page 2012

[7] [7]

J. D. Atkinson,et al., The importance of feldspar for ice nucleation by mineral dust in mixed- phase clouds.Nature498(7454), 355–358 (2013)

work page 2013

[8] [8]

Yakobi-Hancock, L

J. Yakobi-Hancock, L. Ladino, J. Abbatt, Feldspar minerals as efficient deposition ice nuclei. Atmos. Chem. Phys.13(22), 11175–11185 (2013)

work page 2013

[9] [9]

A. D. Harrison,et al., Not all feldspars are equal: a survey of ice nucleating properties across the feldspar group of minerals.Atmos. Chem. Phys.16(17), 10927–10940 (2016)

work page 2016

[10] [10]

A. D. Harrison,et al., The ice-nucleating ability of quartz immersed in water and its atmospheric importance compared to K-feldspar.Atmos. Chem. Phys.19(17), 11343–11361 (2019)

work page 2019

[11] [11]

Kiselev,et al., Active sites in heterogeneous ice nucleation—the example of K-rich feldspars

A. Kiselev,et al., Active sites in heterogeneous ice nucleation—the example of K-rich feldspars. Science355(6323), 367–371 (2017). 13

work page 2017

[12] [12]

T. F. Whale,et al., The role of phase separation and related topography in the exceptional ice-nucleating ability of alkali feldspars.Phys. Chem. Chem. Phys.19(46), 31186–31193 (2017)

work page 2017

[13] [13]

M. A. Holden,et al., High-speed imaging of ice nucleation in water proves the existence of active sites.Sci. Adv.5(2), eaav4316 (2019)

work page 2019

[14] [14]

A. Soni, G. Patey, Simulations of water structure and the possibility of ice nucleation on selected crystal planes of K-feldspar.J. Chem. Phys.150(21) (2019)

work page 2019

[15] [15]

E. Pach, A. Verdaguer, Pores dominate ice nucleation on feldspars.J. Phys. Chem. C123(34), 20998–21004 (2019)

work page 2019

[16] [16]

A. A. Kiselev,et al., Effect of chemically induced fracturing on the ice nucleation activity of alkali feldspar.Atmos. Chem. Phys.21(15), 11801–11814 (2021)

work page 2021

[17] [17]

Kumar, A

A. Kumar, A. K. Bertram, G. N. Patey, Molecular simulations of feldspar surfaces interact- ing with aqueous inorganic solutions: interfacial water/ion structure and implications for ice nucleation.ACS Earth Space Chem.5(8), 2169–2183 (2021)

work page 2021

[18] [18]

Keinert, K

A. Keinert, K. Deck, T. Gaedeke, T. Leisner, A. A. Kiselev, Mechanism of ice nucleation in liquid water on alkali feldspars.Faraday Discuss.235, 148–161 (2022)

work page 2022

[19] [19]

Franceschi,et al., How water binds to microcline feldspar (001).J

G. Franceschi,et al., How water binds to microcline feldspar (001).J. Phys. Chem. Lett.15(1), 15–22 (2023)

work page 2023

[20] [20]

Dickbreder,et al., Atomic structure and water arrangement on K-feldspar microcline (001)

T. Dickbreder,et al., Atomic structure and water arrangement on K-feldspar microcline (001). Nanoscale16(7), 3462–3473 (2024)

work page 2024

[21] [21]

P. M. Piaggi, J. Weis, A. Z. Panagiotopoulos, P. G. Debenedetti, R. Car, Homogeneous ice nucleation in an ab initio machine-learning model of water.Proc. Natl. Acad. Sci. U. S. A. 119(33), e2207294119 (2022)

work page 2022

[22] [22]

A. Soni, G. Patey, Using machine learning with atomistic surface and local water features to predict heterogeneous ice nucleation.J. Chem. Phys.160(12) (2024). 14

work page 2024

[23] [23]

A. Soni, G. Patey, Unraveling the mechanism of ice nucleation by mica (001) surfaces.J. Phys. Chem. C125(48), 26927–26941 (2021)

work page 2021

[24] [24]

A. Soni, G. Patey, How microscopic features of mineral surfaces critically influence heteroge- neous ice nucleation.J. Phys. Chem. C125(19), 10723–10737 (2021)

work page 2021

[25] [25]

Vonnegut, The nucleation of ice formation by silver iodide.J

B. Vonnegut, The nucleation of ice formation by silver iodide.J. Appl. Phys.18(7), 593–595 (1947)

work page 1947

[26] [26]

M. Chen, L. Tan, H. Wang, L. Zhang, H. Niu, Exploring homogeneous ice nucleation through molecular dynamics simulations with ab initio accuracy.Phys. Rev. B:Condens. Matter Mater. Phys.111(13), 134116 (2025)

work page 2025

[27] [27]

P. J. Steinhardt, D. R. Nelson, M. Ronchetti, Bond-orientational order in liquids and glasses. Phys. Rev. B:Condens. Matter Mater. Phys.28(2), 784 (1983)

work page 1983

[28] [28]

P. M. Piaggi, A. Selloni, A. Z. Panagiotopoulos, R. Car, P. G. Debenedetti, A first-principles machine-learning force field for heterogeneous ice nucleation on microcline feldspar.Faraday Discuss.249, 98–113 (2024)

work page 2024

[29] [29]

W. Zhou, P. M. Piaggi, Data from ”The (110) surface is the key to the high ice nucleation efficiency of potassium feldspar”, Zenodo (2025), doi:10.5281/zenodo.17610332,https: //doi.org/10.5281/zenodo.17610332. Acknowledgments We are grateful for the computational resources provided on the Hyperion cluster at the Donostia International Physics Center (DIPC...

work page doi:10.5281/zenodo.17610332 2025